US8204069B2 - Systems and methods for queue management in packet-switched networks - Google Patents
Systems and methods for queue management in packet-switched networks Download PDFInfo
- Publication number
- US8204069B2 US8204069B2 US12/691,310 US69131010A US8204069B2 US 8204069 B2 US8204069 B2 US 8204069B2 US 69131010 A US69131010 A US 69131010A US 8204069 B2 US8204069 B2 US 8204069B2
- Authority
- US
- United States
- Prior art keywords
- queues
- priority
- packets
- class
- queue
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000001514 detection method Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 description 57
- 238000007726 management method Methods 0.000 description 26
- 239000008186 active pharmaceutical agent Substances 0.000 description 14
- HRULVFRXEOZUMJ-UHFFFAOYSA-K potassium;disodium;2-(4-chloro-2-methylphenoxy)propanoate;methyl-dioxido-oxo-$l^{5}-arsane Chemical compound [Na+].[Na+].[K+].C[As]([O-])([O-])=O.[O-]C(=O)C(C)OC1=CC=C(Cl)C=C1C HRULVFRXEOZUMJ-UHFFFAOYSA-K 0.000 description 13
- 238000010586 diagram Methods 0.000 description 8
- 238000012913 prioritisation Methods 0.000 description 8
- 229920000331 Polyhydroxybutyrate Polymers 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 230000006399 behavior Effects 0.000 description 5
- 230000001934 delay Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- RGNPBRKPHBKNKX-UHFFFAOYSA-N hexaflumuron Chemical compound C1=C(Cl)C(OC(F)(F)C(F)F)=C(Cl)C=C1NC(=O)NC(=O)C1=C(F)C=CC=C1F RGNPBRKPHBKNKX-UHFFFAOYSA-N 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000010397 one-hybrid screening Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000005641 tunneling Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
- H04L47/62—Queue scheduling characterised by scheduling criteria
- H04L47/6215—Individual queue per QOS, rate or priority
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/32—Flow control; Congestion control by discarding or delaying data units, e.g. packets or frames
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/50—Queue scheduling
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/90—Buffering arrangements
- H04L49/9084—Reactions to storage capacity overflow
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/02—Standardisation; Integration
- H04L41/0213—Standardised network management protocols, e.g. simple network management protocol [SNMP]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
- H04L41/5003—Managing SLA; Interaction between SLA and QoS
- H04L41/5019—Ensuring fulfilment of SLA
- H04L41/5022—Ensuring fulfilment of SLA by giving priorities, e.g. assigning classes of service
Definitions
- This disclosure relates to methods and systems for queuing traffic in packet-switched networks.
- Packet-switched networks are useful for transmitting packetized data between different network devices.
- Such networks typically include nodes configured to communicate with each other by sending packets over links.
- communications sent from one node must travel through intermediary nodes before reaching the destination node.
- the intermediary nodes are able to decipher routing information contained in the communication packets and forward (i.e., route) the packets in accordance with the information.
- the packets sent by a node may take different network paths to reach the destination node, where the packets are reassembled.
- the routing of packets over packet-switched networks continues to become more complex, especially for networks that support time-sensitive applications (e.g., voice-over-IP) and/or provide quality-of-service (“QoS”) guarantees to customers.
- time-sensitive applications e.g., voice-over-IP
- QoS quality-of-service
- Such networks are able to support different levels of service by prioritizing some communication packets over others. Prioritization of packets for transmission is typically performed in the nodes (e.g., at the routers) of networks, where packets are queued for transmission over a link.
- Priority queuing refers to techniques for queuing packets for transmission based on priority. Many priority queuing techniques separate packets into distinct priority queues based on priority. Such techniques typically decipher priority levels of incoming packets, usually from information contained in the packets, and add the packets to the priority queues, with the highest priority packets being added to the highest priority queues, and so on down through the lowest priority queue. The packets in the different queues are then selected for transmission based on priority, with any queues in the highest priority queue being selected first for transmission. The practice of transmitting higher priority packets ahead of lower priority packets can be referred to as delay priority.
- a queue management system includes a plurality of queues and a priority module configured to assign incoming packets to the queues based on priorities associated with the incoming packets.
- the priority module is further configured to drop at least one of the packets already contained in the queues.
- the priority module is configured to operate across multiple queues when determining which of the packets contained in the queues to drop.
- a queue management system includes a plurality of queues and a priority module configured to assign incoming packets to the queues based on priorities associated with the incoming packets.
- the priority module is further configured to drop one or more of the incoming packets before the number of packets already contained in the queues has reached a maximum occupancy.
- the priority module is configured to operate across multiple queues when determining which of the incoming packets to drop.
- a hybrid queue management system includes classes, queues within each of the classes, and priority levels associated with the queues.
- a classifier is configured to selectively assign incoming packets to the classes.
- a priority module is provided for each of the classes and is configured to assign the incoming packets within each of the classes to the queues based on priorities associated with the incoming packets.
- the priority module is further configured to drop one or more of the incoming packets before the number of packets already contained in the queues has reached a maximum occupancy.
- the priority module is configured to operate across multiple queues when determining which of the incoming packets to drop.
- the system further includes a combination of one or more priority schedulers and class schedulers. The combination is configured to selectively schedule the packets contained in the queues for transmission based on the classes and the priority levels.
- a hybrid queue management system includes classes, queues within each of the classes, and priority levels associated with the queues.
- a classifier is configured to selectively assign incoming packets to the classes.
- a priority module is provided for each of the classes and is configured to assign the incoming packets within each of the classes to the queues based on priorities associated with the incoming packets.
- the priority module is further configured to drop at least one of the packets already contained in the queues.
- the priority module is configured to operate across multiple queues when determining which of the packets contained in the queues to drop.
- the system further includes a combination of one or more priority schedulers and class scheduler. The combination is configured to selectively schedule the packets contained in the queues for transmission based on the classes and the priority levels.
- the priority module is configured to determine a group of the queues based on priority levels associated with the queues.
- the group includes a first queue having a first priority levels and each of the queues having higher priority levels than the first priority level.
- the priority module is configured to operate on the group of queues to determine which packet to drop. Methods for queue management are also provided.
- FIG. 1 is a block diagram illustrating an example of a packet-switched network having an exemplary queue management system implemented therein, according to one embodiment.
- FIG. 2 is a block diagram illustrating one hybrid embodiment of the queue management system of FIG. 1 .
- FIG. 3 is a block diagram illustrating another hybrid embodiment of the queue management system of FIG. 1 .
- FIG. 4 is a block diagram illustrating yet another embodiment of the queue management system of FIG. 1 .
- FIG. 5 is a flowchart illustrating an example of a method for queuing traffic using the exemplary queue management systems of FIGS. 2 and 3 .
- Systems and methods for managing queues in packet-switched networks are disclosed. More specifically, the systems and methods (collectively the “system”) provide for queue management that can drop packets from priority queues based on the priority levels of the packets.
- the system is able to drop incoming packets and/or packets already contained in the queues.
- the system is capable of operating across a set of the queues, including a group of multiple queues.
- priority levels may be associated with the queues, and the system may consider occupancy levels of both a particular queue and all queues having higher priority levels than the particular queue.
- Groups of the priority queues may be dynamically determined based on priority levels of incoming packets.
- the queue or queues within the groups are then considered for drop operations. For example, the occupancy levels of the queues in the group may be considered to help determine whether to drop a packet and/or which packet to drop.
- hybrid embodiments of the system are capable of balancing transmission delay priority with traffic flow fairness, which balance is especially helpful for stabilizing low priority traffic.
- Transmission delay priority refers to the prioritization of high priority packets contained in priority queues for transmission
- flow fairness refers to maintaining a fair traffic flow between different classes and/or flows of traffic.
- Hybrid embodiments of the system may use both class-based and priority-based queuing processes to provide an enhanced beneficial balance between transmission delay priorities and flow fairness so that higher-priority traffic is prioritized for transmission while the flow of lower priority traffic is considered. As a result, overall congestion is reduced while the available bandwidth is used more efficiently. Moreover, the flow of lower priority traffic can be guaranteed while still prioritizing higher priority traffic.
- incoming packets may be assigned to priority queues based on the priority levels associated with the packets.
- a priority module is configured to drop packets from the queues while operating across a group of multiple queues. Accordingly, packets may be dropped from lower priority queues in a manner that considers packets of higher priority queues. In this manner, embodiments of the system are able to avoid congestion by selecting packets to drop based on priority, which process may be referred to as drop priority.
- priority queues and drop priority may be combined with class-based partitioning to provide a balance between flow fairness and transmission delay priority.
- incoming packets may be classified and assigned to distinct classes. Within each of the classes, packets may be assigned to priority queues based on priorities of the packets. Combinations of one or more class-based schedulers and priority schedulers may be configured to determine the order in which packets in the queues will be selected for transmission.
- the system is capable of supporting QoS and/or CoS guarantees for various levels of service. Accordingly, the system is capable of supporting a wide range of applications, including time-sensitive applications such as voice-over-IP (“VoIP”), without unduly delaying the transmission of packets that may not be labeled as time sensitive.
- VoIP voice-over-IP
- FIG. 1 is a block diagram illustrating an example of a packet-switched network 110 having an exemplary queue management system 100 (also referred to simply as “the system 100 ”) implemented therein, according to one embodiment.
- System 100 may be implemented using a variety of hardware approaches known to those of ordinary skill in the art, and the instructions for implementation may be stored using a computer-readable medium.
- the packet-switched network 110 (also referred to simply as “the network 110 ”) may include nodes 130 - 1 through 130 - n (collectively “the nodes 130 ”) connected by links 140 - 1 through 140 - n (collectively “the links 140 ”).
- the nodes 130 are configured to transmit packets 150 - 1 through 150 - n (collectively “the packets 150 ”) over the links 140 .
- the packets 150 may take different paths over the network 110 , as determined by routing disciplines implemented in the nodes 130 .
- the system 100 is configured to queue and manage the packets 150 in one or more of the nodes 130 for routing over the links 140 .
- the system 100 may be implemented as instructions stored in a computer-readable medium or media.
- the medium or media may be part of or otherwise accessible by one or more of the nodes 130 .
- the instructions may be configured to cause a processor (not shown) of a node 130 to perform any of the steps described herein.
- the instructions may be in the form of software and/or firmware applications, or in any other suitable form.
- the system 100 comprises computer-readable instructions configured to cause a node 130 (e.g., a router) to perform any of the predefined queue management processes described below.
- system 100 While an exemplary implementation of the system 100 is shown in FIG. 1 , those skilled in the art will recognize that the exemplary components illustrated in the Figure are for purposes of illustration and not intended to be limiting. Indeed, those skilled in the art will recognize that other alternative network environments may be used in other embodiments. Further, the system 100 may be implemented in one or more of the nodes 130 of the network 110 . Each element of the packet-switched network 110 shown in FIG. 1 will now be discussed in greater detail.
- the packet-switched network 110 may be in the form of a wide variety of different types and/or configurations of networks capable of transmitting packets 150 . More specifically, the packet-switched network 110 may include any suitable technology or technologies (e.g., routing technologies) that enable transmission of packets between the nodes 130 .
- suitable technology or technologies e.g., routing technologies
- the packet-switched network 110 may be configured to transmit packets 150 in accordance with known data structures, network layers, and network protocols, including, but not limited to, transmission control protocol (“TCP”), Internet protocol (“IP”) (e.g., version four and version six), user datagram protocol (“UDP”), real-time transport protocol (“RTP”), utilizing directional antennae in ad hoc networks (“UDAAN”), X.25, asynchronous transfer mode (“ATM”), multi-protocol label switching (“MPLS”), point-to-point protocol (“PPP”), address resolution protocol (“ARP”), synchronous optical networking (“SONET”), integrated services digital network (“ISDN”), and open systems interconnection (“OSI”) layers and protocols, to name a few.
- TCP transmission control protocol
- IP Internet protocol
- RTP real-time transport protocol
- UDAAN ad hoc networks
- ATM asynchronous transfer mode
- MPLS multi-protocol label switching
- PPP point-to-point protocol
- ARP address resolution protocol
- the network 110 may be configured to queue and transmit packets 150 having different protocols and/or OSI layers. As discussed below, the packets 150 may be configured to allow the network 110 to determine and compare priority levels between packets 150 of different protocols and layers. For example, IP packets may be mapped to UDAAN packets in a manner that allows the system 100 to determine the priority of the packets 150 in either protocol.
- the packet-switched network 110 may be configured to transmit packets 120 in accordance with a wide variety of prioritization techniques.
- the packet-switched network 110 may be configured to identify priority labels contained in the packets 150 and prioritize the packets 150 for transmission in accordance with the identified priority levels. This allows the network 110 to support quality-of-service (“QoS”) applications and services, including those that require transmission of time-sensitive data.
- QoS quality-of-service
- the network 110 can transmit high priority network traffic with reduced latency and packet loss. Priority labels will be discussed in further detail below in relation to the packets 150 .
- the network 110 may be configured to transmit packets 150 based on both priority and class.
- the system 100 may be configured to classify packets 150 into different categories (e.g., classes), based on any suitable characteristic (e.g., protocol type) of the packets 150 .
- the classes may be predefined.
- the network 110 may be preconfigured to allocate overall resources (e.g., bandwidth) among the different classes to promote flow fairness, as described below.
- the network 110 may be configured with some form of admission controls for controlling the amount of traffic injected into the network 110 .
- the admission controls may be configured to limit the amount of certain types of packets 150 (e.g., packets 150 having a high-priority label) injected in the network 110 .
- the packet-switched network 110 may use any known or not yet developed communication medium or media for the transmission of packets 150 between the nodes 130 .
- the links 140 may comprise any known medium or media capable of carrying the packets 150 between the nodes 130 of the network 110 .
- the links 140 may be in the form of wireless connections, including, but not limited to radio frequency connections (e.g., cellular connections), infrared connections, or any other known type of wireless connection.
- the links 140 may be in the form of non-wireless connections such as telephone lines, cable lines, fiber lines, etc.
- the packets 150 may be in the form of any known or not yet developed packet structures capable of being transmitted over the network 110 .
- the packets 150 may conform to any of the network protocols mentioned above, including TCP, IP, ARP, UDP, and UDAAN.
- the packets 150 may conform to any layer or protocol of the open systems interconnection (“OSI”) model, which model is well known to those skilled in the art.
- OSI open systems interconnection
- the packets 150 incoming to the nodes 130 are at the network or sub-network layer (e.g., a sub-network layer in the form of UDAAN) of the OSI model, as will be understood by those skilled in the art.
- the packets 150 may be at other layers of the OSI model, including the data link layer and the physical layer.
- the packets 150 include priority labels indicative of different priority levels. This allows the system 100 to prioritize packets 150 for transmission based on their priority labels, as described below.
- the priority labels may be contained in one or more fields of the packets 150 (e.g., in header fields) and may be in any known or not yet developed form.
- the priority labels may be contained in type-of-service (“ToS”), differentiated services (“DS”), explicit congestion notification (“ECN”), and/or per-hop forwarding (“PHB”) fields, which are well-known in the field.
- ToS type-of-service
- DS differentiated services
- ECN explicit congestion notification
- PHB per-hop forwarding
- the priority label is contained in the DS field, and packets 150 are marked with a DS Codepoint (“DSCP”), which may be used by the nodes ( 130 ) to map the packets 150 to different per-hop forwarding behaviors (“PHBs”).
- DSCP DS Codepoint
- PHB per-hop forwarding behaviors
- a PHB defines a particular service level by providing a coded description of externally observable forwarding behavior. Higher priority packets correspond to PHBs that are preferable in terms of both bandwidth and delay, while lower priority packets are relegated to slower services.
- the system 100 may use the PHBs to identify priority levels of packets 150 for queue management purposes.
- the PHBs may be of any form known in the art.
- the PHBs are configured to be at least partially backwards compatible with ToS definitions by including Class Selector Codepoints, which are known to those skilled in the art and are useful for ensuring forwarding compatibility with ToS.
- the packets 150 are in Internet protocol (“IP”) form and use the ToS or DS fields to indicate priority level.
- IP Internet protocol
- OBS optical burst switching
- Updated standards may also be used, including the six-bit differentiated service (“DS”) and two-bit explicit congestion notification (“ECN”) fields.
- DS field is an unstructured six-bit field in which DSCPs are encoded, and is capable of conveying sixty-four distinct codepoints.
- IP packets 150 useful for network-layer routing such as “open shortest path first” (“OSPF”) and “distance vector multicast routing protocol” (“DVMRP”), may be encoded with an “Internetwork Control” DSCP.
- OSPF open shortest path first
- DVMRP distance vector multicast routing protocol
- IP control traffic may contain DSCPs that map to a lower priority.
- Internet group management protocol (“IGMP”) packets may be encoded with the Class Selector Codepoint equivalent to the ToS priority precedence.
- IP packets 150 of type “network time protocol” (“NTP”) and “Internet control message protocol” (“ICMP”) may be encoded with “low delay” and “high throughput” DSCPs, respectively.
- NTP network time protocol
- ICMP Internet control message protocol
- simple network management protocol (“SNMP”) traffic should default to a “low delay” DSCP, but may be elevated to a higher priority DSCP (e.g., “Internetwork Control”).
- General user traffic may be encoded with any DSCP other than that indicating “Network Control” (which may be reserved for sub-network control traffic) and “Internetwork Control.”
- the network 110 may be configured to process DS fields that may map to one or more predetermined priority levels.
- the DS field may define a default DSCP that maps to a “best-effort” forwarding priority level (equivalent to known ToS “Routine” service).
- Non-IP packets may or may not have a DS field. Those with a DS field can be handled similarly to IP packets with a DS field. Those without a DS filed may be treated as if they had a default DSCP, or an implicit mapping may be predefined.
- IP packets 150 may be mapped to UDAAN packets 150 in a manner that allows the system 100 to use the UDAAN packets 150 to identify priority levels.
- UDAAN packets 150 may be marked differently based on their contents. For example, a UDAAN control packet 150 typically does not include a field in which to encode a DSCP. Therefore, UDAAN control packets 150 may be implicitly mapped to a “Network Control” DSCP. This DSCP should map to the highest priority forwarding service for ensuring maintenance of network connectivity and performance.
- UDAAN data packets 150 typically include a ToS field in the header. Accordingly, the system 100 may be configured to determine queuing behaviors of UDAAN data packets 150 by interpreting the ToS field as a DS field. This includes determining a priority level of a UDAAN packet 150 .
- packet protocols that do not include a DS field may be implicitly mapped to a predetermined priority level.
- the system 100 may be configured to map ARP packets to a priority level associated with the Class Selector Codepoint equivalent of the ToS “precedence” priority, which is known in the art.
- the system 100 may be configured to map certain types of packets to a predefined default DSCP. In one embodiment, for example, the system 100 is configured to map all packets not of type IP, ARP, or UDAAN to the default DSCP. Moreover, while it is preferable that incoming packets 150 contain priority information, the system 100 may be configured to handle packets 150 not having priority labels by assigning such packets 150 a default DSCP.
- the explicit congestion notification (“ECN”) field may be used to vary the transmission of a packet flow based on the data contained in the ECN field, as known to those skilled in the art. For example, the transmission rate may be reduced when the ECN field indicates that the associated packet 150 has experienced congestion.
- the packet 150 may be marked as having experienced congestion under a wide variety of conditions, including when the system 100 has chosen the packet 150 to be dropped when a queue is not yet full.
- the ECN field may be used to mark packets 150 of different protocols.
- the ToS field of UDAAN packets 150 may be treated as an ECN field and used to mark the UDAAN packets 150 as having experienced congestion.
- the network 110 may support ECN tunneling, which is known in the art.
- the packets 150 may include information or exhibit other characteristics that are observable by the system 100 .
- the system 100 may be configured to use any such characteristics to classify packets 150 into distinct groups (i.e. classes).
- classes are useful for promoting flow fairness by allocating a portion of overall network resources to each class. For example, overall bandwidth may be shared among several different predefined classes.
- classes are defined according to packet protocols.
- protocols generally known as having “elastic” behavior may comprise one class while protocols generally known as having “inelastic” behavior may be classified in another class.
- the term elastic refers to transmission protocols that adjust transmission rates based on network congestion. To illustrate, an elastic protocol will slow its transmission rate when signs of network congestion are observed (e.g., when one or packets 150 are dropped from a queue or marked in the ECN field as having experienced congestion).
- the term inelastic refers to transmission protocols that do not vary their transmission speeds even when congestion is detected. As known to those skilled in the art, inelastic protocols may tend to choke out elastic protocols. Accordingly, the system 100 is capable of using classes to protect elastic protocol traffic from being overrun by inelastic protocol traffic.
- the nodes 130 may comprise any device or devices capable of routing the packets 150 over the links 140 .
- the nodes 130 may be in the form of routers, switches, gateways, and the like.
- the nodes 130 may include any hardware, firmware, and/or software useful for implementing and operating the system 100 .
- the nodes 130 are capable of routing packets 150 at the network, sub-network, data link, and/or physical layers of the OSI model.
- the nodes 130 may be configured to both encapsulate packets 150 into lower layers of the OSI model, as well as extract higher-layer packets 150 embedded in lower-layer packets 150 , as known to those skilled in the art. Accordingly, the nodes 130 may receive packets 150 at the network or sub-network layers and encapsulate those packets 150 into link and/or physical layers for transmission over the links 140 . The nodes 130 are configured to queue the packets 150 for submission to a hardware interface to the links 140 . The hardware interface transmits the packets 150 over the links 140 to other nodes 140 .
- the queue management system 100 is configured to queue incoming packets 150 for transmission over the links 140 .
- node 130 - 1 may receive incoming packets 150 over one or more links 140 (e.g., link 140 - 1 ).
- the queue management system 100 queues the received packets 150 for transmission based on the priority levels associated with the packets 150 .
- the system 100 may also queue the packets 150 based on both priority levels and classes associated with the packets 150 .
- FIGS. 2 , 3 , and 4 each illustrate an exemplary embodiment of the system 100 .
- FIG. 2 Exemplary Hybrid System
- FIG. 2 is a block diagram of an exemplary queue management system 200 (also referred to as “the system 200 ”), according to one embodiment.
- the system 200 includes a classifier 210 , which is configured to assign incoming packets 150 to predetermined classes 220 - 1 through 220 - k (collectively “the classes 220 ”).
- the packets 150 may be classified based on any observable packet or traffic flow characteristic(s).
- the classifier 210 is configured to classify the packets 150 based on the type of transmission protocol associated with the packets 150 .
- one class 220 - 1 may include any elastic transmission protocols (e.g., TCP), while another class 220 - n may include packets conforming to any inelastic protocol (e.g., UDP).
- the classes 220 may be predefined in any suitable manner.
- the classes 220 include a class for elastic packets, a class for inelastic packets, and a class for control packets.
- Each of the classes 220 may be allocated a share of overall network resources, such as overall bandwidth of a particular link 140 .
- the control class may be allocated ten percent of available bandwidth, and each of the remaining classes may evenly divide the remaining bandwidth.
- the system 200 provides flow fairness between the classes 220 .
- shares of bandwidth are reserved for low priority classes 220 so that prioritized traffic does not completely stop transmission of low priority traffic.
- Classes 220 may be used together with prioritization of packets 150 in order to strike a balance between priority transmissions and flow fairness.
- the allocations of network resources may be flexible. In particular, temporary adjustments to the allocations may be made to handle temporary needs. For example, when the amount of elastic traffic exceeds the amount of bandwidth allocated for the elastic class, additional bandwidth may be temporarily borrowed, if available, from other classes (e.g., the inelastic class) to meet the need. This provides dynamic flexibility to meet the needs of changing distributions of traffic types.
- the system 200 may prioritize the packets 150 within each class 220 , independently of packets 150 in other classes 220 .
- the system 200 includes priority module 230 - 1 through 230 - k (collectively “the priority modules 230 ”), with one priority module 230 - 1 per class 220 .
- the priority modules 230 are configured to assign packets 150 into one or more priority queues 240 - 1 through 240 - n (collectively “the priority queues 240 ”) within each class 220 .
- the priority module 230 may assign (e.g., add) the packets 150 to priority queues 240 using priority labels associated with the packets 150 , including any form of priority labels described above. As previously described, the packets 150 may include a priority label in one or more fields. The priority module 230 is able to access the field(s) and use the priority label contained therein to determine the pre-assigned priority level of any particular packet 150 . For packets 150 that do not have a pre-assigned priority level, the priority module 230 may assign such packets 150 to a predefined default priority queue 240 . In this manner, the system 200 is able to handle both packets 150 having priority labels and packets 150 without any pre-assigned priority levels.
- the priority module 230 employs the functionalities of Random Early Detection (“RED”) to add and drop packets to/from the priority queues 240 .
- RED is an implicit congestion avoidance mechanism that drops packets during periods of network overload in a fair manner, as opposed to standard tail-dropping of packets.
- RED drops packets stochastically according to the average queue length such that the probability of any particular packet flow having a packet dropped is proportional to the size of the flow. More specifically, RED operates by occasionally dropping packets entering a queue when the average queue size exceeds a minimum threshold (e.g., a predefined number of packets 150 ). The exact drop probability is a linear function of average queue size.
- RED tends to alleviate some negative effects of tail dropping by randomly spreading packet drops across the queue
- RED has not in the past dropped packets 150 based on priority levels because RED has previously only been applied to priority queues 240 on an individual basis.
- the priority module 230 extends the capabilities of RED by considering the priority levels of packets 150 during drop operations.
- the extended version of RED may be referred to as priority RED (“P-RED”).
- P-RED priority RED
- each of the priority modules 230 is able to consider which packets 150 to drop based on priority.
- the priority module 230 is configured to consider packet occupancy levels for groups of multiple priority queues 240 for purposes of drop operations.
- the priority module 230 is configured to consider the packets 150 in (i.e., the occupancy level of) a particular priority queue 240 , as well as any packets 150 in (i.e., occupancy levels of) the priority queues 240 having a higher priority level than the particular priority queue 240 .
- a drop operation for the lowest priority queue 240 in a class 220 would consider the occupancy levels in every priority queue 240 in the class 220 .
- a drop operation for the highest priority queue 240 in the class 220 would consider only the occupancy level of the highest priority queues 240 and ignore the lower priority queues 240 in the class 220 .
- the priority module 230 is able to drop packets 150 early, before they are forced to be subjected to lengthy queue delays due to prioritization of higher priority queues 240 , or before tail dropping of packets begins due to a maximum aggregate occupancy threshold being reached.
- the priority module 230 is not limited to determining only whether other packets 150 having the same level of priority are in the corresponding priority queue 240 of the class 220 . Rather, the priority module 230 also checks for packets 150 of higher priority that are in the higher priority queues 240 of the class 220 .
- the priority module 230 is able to recognize potential transmission delays for low priority packets 150 due to the presence of high priority packets 150 in the priority queues 240 . Low priority packets 150 may then be dropped to reduce congestion in the low priority queues 240 .
- the priority modules 230 provide the benefits of RED extended to operate across groups of multiple priority queues 240 for purposes of packet drop functions.
- the groups are dynamic and may be determined based on the priorities of incoming packets 150 .
- the groups may include one or more of the overall number of priority queues 240 .
- the priority drop functions of the priority module 230 allow low priority packets 150 to consider the higher priority packets 150 within the same class 220 . Because low priority packets 150 are not blind to higher priority packets 150 , the system 200 is able to aggressively drop packets 150 to reduce congestion in low priority queues 240 and beneficially enhance overall system performance based on the available bandwidth.
- the packets 150 not dropped from the priority queues 240 generally travel through the queues 240 in any manner known to those skilled in the art, including FIFO. Packets 150 in the packet queues 240 are also subject to being dropped by P-RED operations, as described above.
- the priority queues 240 may be in the form of any suitable data structure(s) (e.g., heaps) and may use any suitable form of memory allocation, including dynamic memory allocation.
- one priority queue 240 is provided for each predetermined priority level. For example, in FIG. 2 , priority queue 240 - 1 may be associated with the lowest priority level, and priority queue 240 - n may be associated with the highest priority level. In other embodiments, multiple priority levels may be grouped into one priority queue 240 .
- the collection of priority queues 240 within a class 220 forms a P-RED group under control of the priority module 230 .
- each class 220 of the system 200 includes a priority scheduler 250 - 1 through 250 - k (collectively “the priority schedulers 250 ”) configured to select the packets 150 in the queues 240 for transmission, based on priority.
- the priority scheduler 250 is configured to prioritize higher priority packets 150 over lower priority packets 150 .
- the priority scheduler 250 may be configured to give highest priority to any packets 150 existing in the priority queue 240 having the highest priority packets 150 (e.g., priority queues 240 - n ). In other words, if any packets 150 exist in the highest priority queue 240 , those packets 150 will be scheduled next, even if packets 150 in lower priority queues 240 are forced to wait.
- the priority schedulers 250 may employ any known scheduling technologies capable of scheduling based on priority.
- each priority scheduler 250 can feed selected packets 150 to a class scheduler 260 .
- the class scheduler 260 is configured to select packets 150 from the priority schedulers 250 and forward the selected packets 150 for transmission based on the class 220 of the packets 150 . Accordingly, the class scheduler 260 is typically able to maintain class-of-service (“CoS”) transmission rates by generally adhering to predetermined per-class allocation of network resources between classes 220 .
- the class-based scheduler 260 may employ any known scheduling technologies capable of scheduling based on class.
- the system 200 effectively provides a hybrid queue management tool that combines priority-based scheduling with class-based scheduling to balance delay priority and flow fairness. Transmission delays are chosen based on the priority levels of packets 150 , yet flow fairness is maintained by the class scheduler 260 . Packets 150 selected by the class scheduler 260 may be forwarded to a single queue (not shown) for a hardware interface (not shown) to a link 140 for transmission over the link 140 .
- the system 200 considers priority levels of packets 150 with respect to adding packets 150 to the queues 240 , dropping packets 150 from the queues 240 , and selecting the order by which packets 150 will be transmitted. In addition to using priority queuing to support QoS requirements, the system 200 keeps prioritization techniques from choking the transmission of low priority packets 150 by scheduling based on classes 220 , as discussed above. Moreover, the system 200 is able to reduce variations in transmission delay for low priority packets 150 by guaranteeing low priority traffic access to a reserved share of overall bandwidth.
- the class scheduler 260 does not observe packet delay priorities across the classes 220 .
- Lower priority packets 150 from one class 220 may be transmitted before packets 150 of a higher priority from a different class 220 .
- the system 200 is particularly well-suited for applications that benefit from balancing flow fairness and delay priority, with flow fairness being preferred over delay priority.
- the system 200 generally tends to favor flow fairness over delay priority because the class scheduler 260 is positioned downstream from the priority schedulers 250 .
- FIG. 3 Another Exemplary Hybrid System
- class schedulers 260 may be positioned upstream from a priority scheduler 250 in order to balance flow fairness and delay priority in a manner that favors delay priority over flow fairness.
- FIG. 3 is a block diagram of another exemplary hybrid queue management system 300 (also referred to as “the system 300 ”), according to one embodiment.
- the system 300 is similar to the system 200 , except that one or more class schedulers 260 may be configured to selectively pull packets 150 from the priority queues 240 .
- a class scheduler 260 may be provided for every set of priority queues 240 having a common priority level.
- class scheduler 260 - 1 may be arranged to pull packets 150 from each priority queue 240 - 1 of every class 220
- class scheduler 260 - n may be arranged to pull packets 150 from each priority queue 240 - n of every class 220
- each class scheduler 260 is able to schedule packets 150 having a common priority level for transmission according to their classes 220 . This keeps packets 150 of all priority levels flowing toward transmission based on predefined per-class bandwidth allocations.
- each of the class schedulers 260 is arranged to feed packets 150 to a priority scheduler 250 , which selects packets 150 for transmission based on priority levels.
- packets 150 selected by the particular class scheduler 260 - n associated with the particular priority queues 240 - n containing the highest priority packets 150 will be prioritized over other packets 150 selected by other class schedulers 260 (e.g., class scheduler 260 - 1 ).
- the system 300 is able to balance flow fairness and delay priority.
- the system 300 provides a balance that tends to favor delay priority over flow fairness.
- the system 300 is well-suited for applications the may benefit from a balance and that also prefer to favor delay priority over flow fairness. Examples of such applications include any network that supports guaranteed QoS levels for transmissions of critical and/or time-sensitive data types (e.g., military radio and VoIP).
- FIG. 4 is a block diagram illustrating yet another embodiment of the queue management system of FIG. 1 .
- the system 400 of FIG. 4 includes the priority module 230 , the priority queues 240 , and the priority scheduler 250 described above in relation to FIG. 2 .
- the system 400 may be implemented in a node 130 without being combined with a class-based component. This may be preferable in networks 110 supporting applications that require strict observance of delay priority at the expense of flow fairness. Even without class-based scheduling, the system 400 is able to reduce the extent to which low priority traffic may experience congestion and variable latencies. This benefit is provided at least in part by the priority module 230 , which is configured to consider priority of packets 150 for drop operations by operating across dynamic groups of priority queues 240 , as discussed above.
- FIG. 5 is a flowchart illustrating an example of a method for queuing traffic using the exemplary queue management systems 200 and 300 . While FIG. 5 illustrates exemplary steps according to one embodiment, other embodiments may omit, add to, and/or modify any of the steps shown in FIG. 5 .
- Step 510 begins by assigning incoming packets 150 into classes 220 at step 510 .
- Step 510 may be performed by the classifier 210 in any of the ways described above.
- Step 520 the packets 150 are assigned to priority queues 240 within each class 220 , based on priority levels associated with the packets 150 .
- Step 520 includes adding packets to the priority queues 240 in any of the ways described above.
- Step 530 packet drop operations are performed within each class 220 , based on priority.
- Step 530 may be performed in any of the ways described above, including treating the priority queues 240 in each class 220 as a single P-RED group.
- Step 540 a combination of priority-based scheduling and class-based scheduling is used to select and forward packets 150 from the priority queues 240 for transmission.
- Step 540 may be performed in any of the ways described above.
- the systems and methods disclosed herein provide queue management that supports packet drops based on priority levels of the packets 150 . Further, the systems and methods provide balances between transmission delay priority and traffic flow fairness.
- the systems and methods may combine class-based and priority-based queuing processes to provide unique balances between transmission delay priorities and flow fairness so that higher-priority traffic is prioritized for transmission while the flow of lower priority traffic is considered.
- the flow of lower priority traffic can be guaranteed without overly limiting prioritization of higher priority traffic.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
Description
Claims (21)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US12/691,310 US8204069B2 (en) | 2004-03-25 | 2010-01-21 | Systems and methods for queue management in packet-switched networks |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US55597204P | 2004-03-25 | 2004-03-25 | |
US11/087,408 US7680139B1 (en) | 2004-03-25 | 2005-03-23 | Systems and methods for queue management in packet-switched networks |
US12/691,310 US8204069B2 (en) | 2004-03-25 | 2010-01-21 | Systems and methods for queue management in packet-switched networks |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/087,408 Continuation US7680139B1 (en) | 2004-03-25 | 2005-03-23 | Systems and methods for queue management in packet-switched networks |
Publications (2)
Publication Number | Publication Date |
---|---|
US20100118883A1 US20100118883A1 (en) | 2010-05-13 |
US8204069B2 true US8204069B2 (en) | 2012-06-19 |
Family
ID=41819553
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/087,408 Expired - Fee Related US7680139B1 (en) | 2004-03-25 | 2005-03-23 | Systems and methods for queue management in packet-switched networks |
US12/691,310 Expired - Fee Related US8204069B2 (en) | 2004-03-25 | 2010-01-21 | Systems and methods for queue management in packet-switched networks |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US11/087,408 Expired - Fee Related US7680139B1 (en) | 2004-03-25 | 2005-03-23 | Systems and methods for queue management in packet-switched networks |
Country Status (1)
Country | Link |
---|---|
US (2) | US7680139B1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022271247A1 (en) * | 2021-06-26 | 2022-12-29 | Intel Corporation | Predictive queue depth |
Families Citing this family (49)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050152373A1 (en) * | 2004-01-08 | 2005-07-14 | Interdigital Technology Corporation | Packet scheduling in a wireless local area network |
US7680139B1 (en) * | 2004-03-25 | 2010-03-16 | Verizon Patent And Licensing Inc. | Systems and methods for queue management in packet-switched networks |
KR100679013B1 (en) * | 2004-08-11 | 2007-02-05 | 삼성전자주식회사 | Network device and data transmitting method using network device |
US8165144B2 (en) * | 2005-08-17 | 2012-04-24 | Cisco Technology, Inc. | Shaper-scheduling method and system to implement prioritized policing |
CN100563202C (en) * | 2005-09-01 | 2009-11-25 | 华为技术有限公司 | The method of differential service is provided |
US8576846B2 (en) * | 2005-10-05 | 2013-11-05 | Qualcomm Incorporated | Peer-to-peer communication in ad hoc wireless network |
GB0606367D0 (en) * | 2006-03-30 | 2006-05-10 | Vodafone Plc | Telecommunications networks |
US7961745B2 (en) * | 2006-09-16 | 2011-06-14 | Mips Technologies, Inc. | Bifurcated transaction selector supporting dynamic priorities in multi-port switch |
US7990989B2 (en) * | 2006-09-16 | 2011-08-02 | Mips Technologies, Inc. | Transaction selector employing transaction queue group priorities in multi-port switch |
US20080175152A1 (en) * | 2006-12-28 | 2008-07-24 | Nokia Corporation | Service differentiating and overload indication for downlink |
KR20090085906A (en) * | 2008-02-05 | 2009-08-10 | 삼성전자주식회사 | Routing processing system and the control method by priority order of logic interface |
US8406230B2 (en) * | 2008-06-30 | 2013-03-26 | Oracle America, Inc. Formerly Known As Sun Microsystems, Inc. | Method and system for classifying packets in a network interface card and interface for performing the same |
US8780762B2 (en) * | 2009-03-11 | 2014-07-15 | Sony Corporation | Node query in ad hoc home mesh network |
US8761174B2 (en) * | 2009-03-11 | 2014-06-24 | Sony Corporation | Quality of service traffic recognition and packet classification home mesh network |
US8194593B2 (en) * | 2009-03-11 | 2012-06-05 | Sony Corporation | Quality of service architecture for home mesh network |
US8861445B2 (en) * | 2009-03-11 | 2014-10-14 | Sony Cororation | Multi-channel single radio communication in home mesh network |
US20100232400A1 (en) * | 2009-03-11 | 2010-09-16 | Sony Corporation | Virtualizing single radio for multiple wireless interfaces in home mesh network |
US8223786B2 (en) * | 2009-03-11 | 2012-07-17 | Sony Corporation | Quality of service scheduling for home mesh network |
US8005106B2 (en) * | 2009-05-21 | 2011-08-23 | Cisco Technology, Inc. | Apparatus and methods for hybrid fair bandwidth allocation and drop precedence |
US9065779B2 (en) | 2009-06-12 | 2015-06-23 | Wi-Lan Labs, Inc. | Systems and methods for prioritizing and scheduling packets in a communication network |
US8665724B2 (en) | 2009-06-12 | 2014-03-04 | Cygnus Broadband, Inc. | Systems and methods for prioritizing and scheduling packets in a communication network |
US8301805B2 (en) * | 2009-09-15 | 2012-10-30 | Hewlett-Packard Development Company, L.P. | Managing I/O request in a storage system |
WO2011067409A1 (en) | 2009-12-04 | 2011-06-09 | Napatech A/S | An assembly and a method of receiving and storing data while saving bandwidth by controlling updating of fill levels of queues |
US8934341B2 (en) | 2009-12-04 | 2015-01-13 | Napatech A/S | Apparatus and a method of receiving and storing data packets controlled by a central controller |
US8537846B2 (en) * | 2010-04-27 | 2013-09-17 | Hewlett-Packard Development Company, L.P. | Dynamic priority queue level assignment for a network flow |
US8537669B2 (en) * | 2010-04-27 | 2013-09-17 | Hewlett-Packard Development Company, L.P. | Priority queue level optimization for a network flow |
US9262346B2 (en) * | 2010-06-21 | 2016-02-16 | Hewlett Packard Enterprises Development LP | Prioritizing input/outputs at a host bus adapter |
GB2493130B (en) * | 2011-07-11 | 2018-08-01 | Metaswitch Networks Ltd | SIP server overload control |
US20150126211A1 (en) * | 2012-06-06 | 2015-05-07 | Kyocera Corporation | Communication control method and base station |
KR20140011539A (en) * | 2012-07-02 | 2014-01-29 | 한국전자통신연구원 | System and method of virtualization for network application and the apparatus |
US9467388B2 (en) * | 2012-08-29 | 2016-10-11 | Universiteit Gent | Method and device for scheduling data traffic |
US9596182B2 (en) * | 2013-02-12 | 2017-03-14 | Adara Networks, Inc. | Controlling non-congestion controlled flows |
US9092974B2 (en) * | 2013-05-01 | 2015-07-28 | Microsoft Technology Licensing, Llc | Mixing data and control codes on a wireless link |
US9398118B2 (en) * | 2013-09-24 | 2016-07-19 | Innovask, Inc. | Processing apparatus for bus data |
US10356014B2 (en) * | 2013-09-24 | 2019-07-16 | Innovasic, Inc. | Method of processing bus data |
US9923839B2 (en) * | 2015-11-25 | 2018-03-20 | International Business Machines Corporation | Configuring resources to exploit elastic network capability |
US10581680B2 (en) | 2015-11-25 | 2020-03-03 | International Business Machines Corporation | Dynamic configuration of network features |
WO2018002688A1 (en) * | 2016-06-27 | 2018-01-04 | Telefonaktiebolaget Lm Ericsson (Publ) | Head drop scheduler |
US10862617B2 (en) * | 2017-05-30 | 2020-12-08 | Marvell Asia Pte, Ltd. | Flowlet scheduler for multicore network processors |
EP3678339A4 (en) * | 2017-09-29 | 2020-07-15 | Huawei Technologies Co., Ltd. | Message processing method and device |
US20210075734A1 (en) * | 2018-05-31 | 2021-03-11 | Hewlett-Packard Development Company, L.P. | Packets transmissions based on priority levels |
US11350142B2 (en) * | 2019-01-04 | 2022-05-31 | Gainspan Corporation | Intelligent video frame dropping for improved digital video flow control over a crowded wireless network |
WO2020151802A1 (en) * | 2019-01-21 | 2020-07-30 | Telefonaktiebolaget Lm Ericsson (Publ) | Methods and apparatus for packet dropping in a fronthaul network |
CN113162790B (en) * | 2020-01-22 | 2023-10-03 | 华为技术有限公司 | Method, device, equipment and storage medium for adjusting service level |
US11558316B2 (en) * | 2021-02-15 | 2023-01-17 | Mellanox Technologies, Ltd. | Zero-copy buffering of traffic of long-haul links |
US11831553B2 (en) * | 2021-07-16 | 2023-11-28 | Nokia Solutions And Networks Oy | Distinguishing traffic-engineered packets and non-traffic-engineered packets |
US11973696B2 (en) | 2022-01-31 | 2024-04-30 | Mellanox Technologies, Ltd. | Allocation of shared reserve memory to queues in a network device |
US20230344768A1 (en) * | 2022-04-22 | 2023-10-26 | Huawei Technologies Co., Ltd. | System and method for a scalable source notification mechanism for in-network events |
US20240244011A1 (en) * | 2023-01-18 | 2024-07-18 | Mediatek Inc. | Data Sending Control Method and Data Sending Control System Capable of Dynamically Allocating Log Data |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4136399A (en) * | 1977-05-20 | 1979-01-23 | Rca Corporation | Dynamic channel allocation buffer matrix |
US6087298A (en) * | 1996-05-14 | 2000-07-11 | Engelhard Corporation | Exhaust gas treatment system |
US20020027879A1 (en) | 2000-08-10 | 2002-03-07 | Kouji Kuriki | Packet fluctuation absorbing method and apparatus |
US20020048277A1 (en) | 2000-05-01 | 2002-04-25 | Bennett Jon C.R. | Packetized data discard |
US20030165148A1 (en) * | 2002-03-01 | 2003-09-04 | Bishard Clint J. | Queuing closed loop congestion mechanism |
US6643293B1 (en) | 1997-09-05 | 2003-11-04 | Alcatel Canada Inc. | Virtual connection shaping with hierarchial arbitration |
US6680933B1 (en) | 1999-09-23 | 2004-01-20 | Nortel Networks Limited | Telecommunications switches and methods for their operation |
US6865185B1 (en) * | 2000-02-25 | 2005-03-08 | Cisco Technology, Inc. | Method and system for queuing traffic in a wireless communications network |
US6888824B1 (en) * | 2000-10-20 | 2005-05-03 | Cisco Technology, Inc. | Random early detection (RED) algorithm using marked segments to detect congestion in a computer network |
US20050135414A1 (en) * | 1999-07-06 | 2005-06-23 | Alcatel | Scheme for packet allocation in a radiocommunication system |
US6912226B2 (en) * | 2000-03-29 | 2005-06-28 | Alcatel | Method to generate an acceptance decision in a telecommunication system |
US20050157645A1 (en) * | 2004-01-20 | 2005-07-21 | Sameh Rabie | Ethernet differentiated services |
US7073005B1 (en) | 2002-01-17 | 2006-07-04 | Juniper Networks, Inc. | Multiple concurrent dequeue arbiters |
US7680139B1 (en) * | 2004-03-25 | 2010-03-16 | Verizon Patent And Licensing Inc. | Systems and methods for queue management in packet-switched networks |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6430154B1 (en) * | 1999-08-13 | 2002-08-06 | Fujitsu Network Communications, Inc. | Supporting multiple application traffic types over connection oriented networks |
-
2005
- 2005-03-23 US US11/087,408 patent/US7680139B1/en not_active Expired - Fee Related
-
2010
- 2010-01-21 US US12/691,310 patent/US8204069B2/en not_active Expired - Fee Related
Patent Citations (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4136399A (en) * | 1977-05-20 | 1979-01-23 | Rca Corporation | Dynamic channel allocation buffer matrix |
US6087298A (en) * | 1996-05-14 | 2000-07-11 | Engelhard Corporation | Exhaust gas treatment system |
US6643293B1 (en) | 1997-09-05 | 2003-11-04 | Alcatel Canada Inc. | Virtual connection shaping with hierarchial arbitration |
US20050135414A1 (en) * | 1999-07-06 | 2005-06-23 | Alcatel | Scheme for packet allocation in a radiocommunication system |
US6680933B1 (en) | 1999-09-23 | 2004-01-20 | Nortel Networks Limited | Telecommunications switches and methods for their operation |
US6865185B1 (en) * | 2000-02-25 | 2005-03-08 | Cisco Technology, Inc. | Method and system for queuing traffic in a wireless communications network |
US6912226B2 (en) * | 2000-03-29 | 2005-06-28 | Alcatel | Method to generate an acceptance decision in a telecommunication system |
US20020048277A1 (en) | 2000-05-01 | 2002-04-25 | Bennett Jon C.R. | Packetized data discard |
US20020027879A1 (en) | 2000-08-10 | 2002-03-07 | Kouji Kuriki | Packet fluctuation absorbing method and apparatus |
US6888824B1 (en) * | 2000-10-20 | 2005-05-03 | Cisco Technology, Inc. | Random early detection (RED) algorithm using marked segments to detect congestion in a computer network |
US7073005B1 (en) | 2002-01-17 | 2006-07-04 | Juniper Networks, Inc. | Multiple concurrent dequeue arbiters |
US20030165148A1 (en) * | 2002-03-01 | 2003-09-04 | Bishard Clint J. | Queuing closed loop congestion mechanism |
US7362749B2 (en) | 2002-03-01 | 2008-04-22 | Verizon Business Global Llc | Queuing closed loop congestion mechanism |
US20050157645A1 (en) * | 2004-01-20 | 2005-07-21 | Sameh Rabie | Ethernet differentiated services |
US7680139B1 (en) * | 2004-03-25 | 2010-03-16 | Verizon Patent And Licensing Inc. | Systems and methods for queue management in packet-switched networks |
Non-Patent Citations (2)
Title |
---|
"Priority queue", http://en.wikipedia.orgivvikilPriority-queue (last modified Jan. 17, 2005) (2 pages), Jan. 17, 2005. |
"Priority queue", http://en.wikipedia.orgivvikilPriority—queue (last modified Jan. 17, 2005) (2 pages), Jan. 17, 2005. |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022271247A1 (en) * | 2021-06-26 | 2022-12-29 | Intel Corporation | Predictive queue depth |
Also Published As
Publication number | Publication date |
---|---|
US20100118883A1 (en) | 2010-05-13 |
US7680139B1 (en) | 2010-03-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8204069B2 (en) | Systems and methods for queue management in packet-switched networks | |
Semeria | Supporting differentiated service classes: queue scheduling disciplines | |
US6996062B1 (en) | Policy-based weighted random early detection method for avoiding congestion in internet traffic | |
KR100800881B1 (en) | Method and apparatus for supporting qos in residential ethernet system | |
US7983299B1 (en) | Weight-based bandwidth allocation for network traffic | |
US7006437B2 (en) | Scheduling mechanisms for use in mobile ad hoc wireless networks for achieving a differentiated services per-hop behavior | |
US7006440B2 (en) | Aggregate fair queuing technique in a communications system using a class based queuing architecture | |
US7936770B1 (en) | Method and apparatus of virtual class of service and logical queue representation through network traffic distribution over multiple port interfaces | |
US8064344B2 (en) | Flow-based queuing of network traffic | |
US7263063B2 (en) | Per hop behavior for differentiated services in mobile ad hoc wireless networks | |
US7889743B2 (en) | Information dissemination method and system having minimal network bandwidth utilization | |
US8547846B1 (en) | Method and apparatus providing precedence drop quality of service (PDQoS) with class-based latency differentiation | |
WO2007101152A2 (en) | Management of packet congestion in packet switching network | |
US20080080382A1 (en) | Refined Assured Forwarding Framework for Differentiated Services Architecture | |
KR100501717B1 (en) | Method for voice/data transport over UDP/TCP/IP networks using an efficient buffer management | |
US20120176903A1 (en) | Non-uniform per-packet priority marker for use with adaptive protocols | |
US8203956B1 (en) | Method and apparatus providing a precedence drop quality of service (PDQoS) | |
US8379518B2 (en) | Multi-stage scheduler with processor resource and bandwidth resource allocation | |
US20050068798A1 (en) | Committed access rate (CAR) system architecture | |
US7957394B1 (en) | Automatic network switch configuration to support quality of service | |
US11343193B2 (en) | Apparatus and method for rate management and bandwidth control | |
Zoriđ et al. | Fairness of scheduling algorithms for real-time traffic in DiffServ based networks | |
US7061919B1 (en) | System and method for providing multiple classes of service in a packet switched network | |
Astuti | Packet handling | |
Cisco | QC: Quality of Service Overview |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: VERIZON CORPORATE SERVICES GROUP INC., NEW YORK Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:JONES, CHRISTINE E.;TROXEL, GREGORY D.;SIGNING DATES FROM 20050322 TO 20050408;REEL/FRAME:025618/0321 Owner name: BBNT SOLUTIONS LLC, MASSACHUSETTS Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:JONES, CHRISTINE E.;TROXEL, GREGORY D.;SIGNING DATES FROM 20050322 TO 20050408;REEL/FRAME:025618/0321 Owner name: BBN TECHNOLOGIES OPERATING CORP., MASSACHUSETTS Free format text: MERGER;ASSIGNOR:BBNT SOLUTIONS LLC;REEL/FRAME:025618/0369 Effective date: 20051122 Owner name: RAYTHEON BBN TECHNOLOGIES CORP., MASSACHUSETTS Free format text: CHANGE OF NAME;ASSIGNOR:BBN TECHNOLOGIES CORP.;REEL/FRAME:025619/0820 Effective date: 20091027 Owner name: BBN TECHNOLOGIES CORP., MASSACHUSETTS Free format text: CHANGE OF NAME;ASSIGNOR:BBN TECHNOLOGIES OPERATING CORP.;REEL/FRAME:025619/0636 Effective date: 20051201 |
|
AS | Assignment |
Owner name: VERIZON PATENT AND LICENSING INC., NEW JERSEY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:VERIZON CORPORATE SERVICES GROUP INC.;REEL/FRAME:025626/0770 Effective date: 20100112 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
LAPS | Lapse for failure to pay maintenance fees |
Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20200619 |